ABSTRACT Overall Hypothesis: We hypothesize that orally administered genetically modified non-toxigenic C. difficile strains that express antigenic but non-toxic portions of C. difficile toxin A or toxin B will elicit an immune response by the host that will protect against future symptomatic Clostridium difficile infection (CDI) after colonization with the non-toxigenic vaccine strain is lost. Specific Aims: We have identified by restriction endonuclease analysis (REA), specific strains of non- toxigenic C. difficile (NTCD) that were originally isolated from asymptomatically colonized humans and have shown that the strains will successfully, safely, and durably colonize (for a period of weeks to months) and prevent CDI in hamsters challenged with virulent, toxigenic, strains of C. difficile known to cause epidemic disease in humans. We have not detected any serum immune response to the surface proteins of these non- toxigenic strains in hamsters and have found that protection against CDI is correlated with presence of the organism in stool (colonization). We now propose to use these non-toxigenic strains to induce a protective and lasting immune response that will extend protection against CDI beyond the period of colonization. Our Aims are 1. Identify a group of NTCD strains from our library that can be genetically manipulated, 2. Construct recombinant NTCD strains that express antigens from the binding region of toxin A and toxin B, 3. Conduct in vitro analysis of the recombinant non-toxigenic strains to demonstrate stable expression of the toxin A and toxin B vaccine antigens while retaining other phenotypic characteristics, 4. Demonstrate in vivo antibody response to toxin A and toxin B vaccine antigen peptides by colonizing hamsters. 5. Assure that the recombinant NTCD will still provide colonization protection against toxigenic strain challenge, and 6. Demonstrate prevention of CDI in hamsters previously colonized with the NTCD vaccine strains. Background and Significance: CDI rates in hospitals have been rising since the year 2000 and mortality and morbidity is increasing at a rapid rate, particularly among the elderly. Patients in VA hospitals are no exception and in many instances are at higher risk due to advanced age and longer length of stay, particularly in long term care environments. An epidemic strain, BI/NAP1/027, is contributing to the outbreak of CDI. Preliminary Studies: We have found that strains of NTCD that colonize the gut of patients and hamsters are protective against CDI when challenged with toxigenic C. difficile. However, protection is correlated with colonization which lasts only for weeks to months, after which the subject is again vulnerable to CDI. We propose to genetically manipulate these NTCD strains so that they will express peptides from the binding (non- toxic) C-terminus of toxins A and B. To date we have successfully introduced shuttle plasmid pMTL9361 from E. coli into NTCD M3. We have also constructed an expression plasmid carrying the slpA promoter (PslpA) and signal-peptide encoding region (SP) plus the tcdA or tcdB binding regions of toxin A or B, and successfully subcloned the expression plasmid inserts for tcdB into the shuttle plasmid to create a toxin B vaccine plasmid. Research Design and Methods: Our research plan is to identify additional NTCD strains into which we can introduce the pMTL9361 shuttle plasmid from E. coli, to eventually serve as additional recipients for the vaccine shuttle plasmid. We will demonstrate expression of the binding region peptides of toxin A and B in these recombinant NTCD strains, and demonstrate in vitro that we have not lost critical growth and adherence characteristics of the parent NTCD strains. We will then test these strains in the hamster model for colonization and induction of a serum immune response. We will then test the NTCD strains for protection against challenge by toxigenic C. difficile, both during colonization (to assure that colonization protection is not lost) and after colonization has been lost, to determine if the immune response generated by the oral vaccine is protective against toxigenic C. difficile challenge. PUBLIC HEALTH RELEVANCE: Relevance to Veterans Health VA Medical Centers because of their elderly and often chronic patient populations coupled with nursing homes and long-term care facilities have had a disproportionate problem with C. difficile infection (CDI). Well documented outbreaks and endemic CDI have been reported from Cleveland, Richmond, Minneapolis, Tucson, Chicago, and Albany VA. VA Hospital rates of CDI nationally have risen from <4 cases/1000 discharges in 1994 to >12 cases/1000 discharges in 2005, more than tripling previous rates. CDI rates in elderly veterans over the age of 65 are nearly 20/1000 discharges, and directly attributable mortality among patients over age 60 ranges from 5% to 14%, increasing with each decade. Prevention of this disease by a combined colonization protection and vaccine strategy in VA facilities would not only markedly reduce the severe morbidity and mortality of veteran patients, but would also reduce needless prolonged length of stay in hospitals and hospital costs.